How it works
Connect, model, ask, simulate, decide.
Five steps, and the evidence survives all five. Nothing in this sequence asks a government to rebuild its information infrastructure before it can start.
01 / Step
Connect
Whatever exists is connected where it already lives.
A deployment starts with an inventory of what the institution already produces: adopted plans, zoning and land use, the capital programme, budgets, GIS layers, demographic series, asset inventories, environmental readings and any open data already published. Nothing has to be rebuilt or migrated to begin, and the model reports how much of itself each connection fills.
- Documents are read and their statements attached to the areas and problems they concern.
- Geographic services are consumed as sources, not copied into a second system.
- Series that update, such as air quality or temperature, keep their history.
- Every source records its refresh frequency, so an answer can say how current it is.
Model coverage
68%
How much of the model the connected sources fill. It does not measure how accurate or how current any single value is.
- Comprehensive planDOC
- Capital improvement planDOC
- Zoning and land useGIS
- Street networkGIS
- Population by areaSET
- Air quality seriesAPI
- Utility networkGIS
- Daytime populationSET
Declared gaps
- Daytime population is absent, so access analysis covers residents only.
- Utility network is connected for two of five districts.
Product interface concept. Figures are illustrative and describe no real jurisdiction.
02 / Step
Model
Fragments become a structure with sources attached.
The connected material is resolved into one model: territorial areas, services and facilities, infrastructure, projects, indicators and the problems the institution has itself recognised, along with the relationships between them. A project is linked to the problem it addresses. An indicator is linked to the area it describes. What is missing is recorded as a declared gap rather than left as silence.
- Areas, services, infrastructure, projects, indicators and problems, in one schema.
- Each element carries its source, its basis and how current it is.
- Coverage is computed and shown as a figure, not asserted in a sentence.
- Declared gaps are fields in the model, so every downstream answer inherits them.
- AreasTerritorial units and their boundaries
- ServicesFacilities, catchments and coverage
- InfrastructureNetworks, assets and capacity
- ProjectsProgrammed work and its maturity
- ProblemsWhat the institution has recognised
- IndicatorsSeries with source and frequency
Every element also carries: source, basis, refresh frequency, confidence, and the gaps declared against it.
Product interface concept. Figures are illustrative and describe no real jurisdiction.
03 / Step
Ask
A question in plain language, answered from the model.
The question is answered from the connected model rather than from general knowledge, which is what makes the citation possible. The answer names the documents, layers and indicators it used, keeps measured values separate from derived ones and from assumptions, and states where the data runs out. When the sources cannot support a conclusion, that is the answer.
- Citations attach to the claims they support, not to the page.
- Measured, derived and assumed values remain distinguishable downstream.
- Confidence is reported with the reason behind it.
- A gap is returned as a finding, which is often the useful part.
Question
Which neighbourhoods have the largest infrastructure gaps?
Answer
Two districts stand out. Both combine below-median network coverage with above-median population density, and in both cases the gap is already named in the adopted plan without a funded project attached to it.
Evidence
- Network coverage in both districts falls in the lowest quartile of the model.
- Population density in both is above the model median.
- Both appear in the recognised problems list; neither is matched to a programmed project.
Districts flagged
2
Derived
Residents affected
27,600
Derived
Unmatched problems
2
Measured
Sources used in this answer
- Utility and network layersGIS layerPeriodicMeasured
- Population by districtDatasetCensus cycleMeasured
- Adopted plan, infrastructure chapterDocumentOn adoptionMeasured
- Capital improvement programmeDocumentAnnualMeasured
ConfidenceModerate
Network coverage is connected for three of five districts. The two flagged here are inside the connected area; the ranking cannot be extended to the other two.
What the data does not cover
- Network layers are connected for three of five districts.
- Condition data predates the last resurfacing programme, so age is known and current state is not.
Product interface concept. Figures are illustrative and describe no real jurisdiction.
04 / Step
Simulate
Place the intervention and measure what it reaches.
A proposed investment is positioned on the map and evaluated against the model. The catchment is computed by walking the street network outward from the site, so a river, a rail corridor or a motorway costs the site the population on the other side, exactly as it will in practice. Save the result, build the alternative, and the platform sets them against each other on the same measures.
- Isochrones over the real network rather than a radius circle.
- Population reached reported where population data is connected.
- Overlap with existing catchments reported alongside new coverage.
- Two or three scenarios compared on identical indicators.
- Candidate site
- 15 min walking catchment
- Existing facilities
Western district edge, adjacent to an existing transit stop.
Central district, closer to the population centroid.
Population served, 15 min walk
18,40014,900Residents currently unserved
9,1003,600Median walking time
12.4min9.8minOverlap with existing catchments
14%38%
Difference
A reaches 3,500 more residents and 2.5 times as many people currently outside any catchment. B is quicker to reach for those it serves, but most of its catchment is already covered.
TwinGov reports the difference. The choice, and the responsibility for it, stay with the institution.
Product interface concept. Figures are illustrative and describe no real jurisdiction.
05 / Step
Decide
The evidence leaves the platform intact.
A report assembles the map, the figures, the assumptions and the source list into one document, with the declared gaps placed before the conclusions rather than after them. The model itself exports as JSON against a published schema, and geographic layers as standard GeoJSON, so the institution is never dependent on this interface to read its own work.
- A document for a committee file or a funding application.
- Printable, or exported for editing in a word processor.
- Assumptions and gaps travel with the figures, not separately.
- Model and layers export in open formats at any time.
Community facility siting
Comparison of four candidate locations on population reached, coverage added and existing service gaps.
- 1The question and how it was framed
- 2Candidate locationsFour sites, same method
- 3Population reached and coverage added
- 4Assumptions used
- 5Sources
- 6What this analysis does not coverDeclared before conclusions, not after
Product interface concept. Figures are illustrative and describe no real jurisdiction.
Getting started
The first model is built from what you already publish.
An initial model can be assembled from available public and institutional sources before any procurement decision is made, which is usually the fastest way to find out whether the platform is worth the conversation.
01
Scope the sources
We look at what your institution already produces and publishes, and say what a model built from it could and could not answer.
02
Build an initial model
A first version, with its coverage figure and its declared gaps visible from the start rather than presented once they are fixed.
03
Work a real question
One decision your team is actually facing, taken through the model end to end, including the parts where the data does not reach.
What could TwinGov reveal about your community?
Start with the data you already have. We can build an initial model from available public and institutional sources and show where TwinGov can create immediate decision value.